Bars rejuvenating bulges? Evidence from stellar population analysis

We obtained stellar ages and metallicities via spectrum fitting for a sample of 575 bulges with spectra available from the Sloan Digital Sky Survey. The structural properties of the galaxies have been studied in detail in Gadotti (2009b) and the sample contains 251 bulges in galaxies with bars. Using the whole sample, where galaxy stellar mass distributions for barred and unbarred galaxies are similar, we find that bulges in barred and unbarred galaxies occupy similar loci in the age vs. metallicity plane. However, the distribution of bulge ages in barred galaxies shows an excess of populations younger than ~ 4 Gyr, when compared to bulges in unbarred galaxies. Kolmogorov-Smirnov statistics confirm that the age distributions are different with a significance of 99.94%. If we select sub-samples for which the bulge stellar mass distributions are similar for barred and unbarred galaxies, this excess vanishes for galaxies with bulge mass log M < 10.1 M_Sun while for more massive galaxies we find a bimodal bulge age distribution for barred galaxies only, corresponding to two normal distributions with mean ages of 10.4 and 4.7 Gyr. We also find twice as much AGN among barred galaxies, as compared to unbarred galaxies, for low-mass bulges. By combining a large sample of high quality data with sophisticated image and spectral analysis, we are able to find evidence that the presence of bars affect the mean stellar ages of bulges. This lends strong support to models in which bars trigger star formation activity in the centers of galaxies.

An observers view of simulated galaxies: disc-to-total ratios, bars, and (pseudo-)bulges

We use cosmological hydrodynamical simulations of the formation of Milky Way-mass galaxies to study the relative importance of the main stellar components, i.e., discs, bulges, and bars, at redshift zero. The main aim of this work is to understand if estimates of the structural parameters of these components determined from kinematics (as is usually done in simulations) agree well with those obtained using a photometric bulge/disc/bar decomposition (as done in observations). To perform such a comparison, we have produced synthetic observations of the simulation outputs with the Monte-Carlo radiative transfer code SUNRISE and used the BUDDA code to make 2D photometric decompositions of the resulting images (in the i and g bands). We find that the kinematic disc-to-total ratio (D/T) estimates are systematically and significantly lower than the photometric ones. While the maximum D/T ratios obtained with the former method are of the order of 0.2, they are typically >0.4, and can be as high as 0.7, according to the latter. The photometric decomposition shows that many of the simulated galaxies have bars, with Bar/T ratios in the range 0.2-0.4, and that bulges have in all cases low Sersic indices, resembling observed pseudo-bulges instead of classical ones. Simulated discs, bulges and bars generally have similar (g-i) colours, which are in the blue tail of the distribution of observed colours. This is not due to the presence of young stars, but rather to low metallicities and poor gas content in the simulated galaxies, which makes dust extinction low. Photometric decompositions thus match the component ratios usually quoted for spiral galaxies better than kinematic decompositions, but the shift is insufficient to make the simulations consistent with observed late-type systems.